This essay—which summarizes the basic theses of my forthcoming book of the same title—seeks to provide a clear and concise formulation of my provisional conclusions on two key issues in the social sciences: the notions of ideology and class.footnote1 In it, I shall deal successively with four problems: 1. the distinction between superstructure and infrastructure; 2. the relation between determination in the last instance by the economy and the domination of various superstructures; 3. the ‘ideal’ component of social reality, and the distinction between ideological and non-ideological among ideal realities; 4. the role of violence and consent in the power of domination of a class, order, etc. (is there a paradox of ‘legitimacy’ in the birth of classes and the state?). I want to emphasize, before proceeding further, how much I have benefited from my experience—albeit partial—of the wealth of new material in anthropology and history, which is growing all the time. I have only an amateur’s knowledge of the latter discipline, in which my reading has been directed above all towards understanding the problem of the formation of the state and the transformation of class relations. Of course, too, I shall disappoint some readers who would have liked to see a more precise account of the links between my general, abstract positions and this wealth of new anthropological material.

For a Marxist, an enquiry into ideology, the conditions of its formation and transformation, its effects on the motion of societies, should apparently be an enquiry into the relation between infrastructure, superstructures and ideology. Must these realities be baptized ‘instances’ as Althusser does? Should they be considered as ‘levels’ of social reality, as distinctions within social reality which are in some sense substantive, as institutional divides in its substance? I think not. For a society does not have a top or a bottom, or in any real sense levels at all. This is because the distinction between infrastructure and superstructures is not a distinction between institutions. It is essentially a distinction between functions.

What, then, is the meaning of the term ‘infrastructure’? It designates the combination, which exists in all societies, of at least three ensembles of material and social conditions, which allow members of a society to produce and reproduce material means of their social existence. These ensembles are:

1. The determinate ecological and geographical conditions within which a society exists, and from which it extracts its material means of existence. 2. The productive forces: that is to say, the material and intellectual means which members of a society set in motion, via various ‘labour processes’, in order to act on nature and extract from it their means of existence—thus transforming and ‘socializing’ it.

3. The social relations of production. That is, those relations, whatsoever they be, which assume one or more of the following functions: a) determining the social form of access to resources and control over the means of production; b) redistributing the labour power of society’s members between the various labour processes which provide the latter’s material basis, and organizing these processes; c) determining the social form of redistribution of the products of individual or collective labour and, as a result, the form of their circulation or non-circulation.

Let us recall that, in the strict sense, what Marx termed the economic structure of society is only the social relations of production: ‘the totality of these relations of production constitutes the economic structure of society’.footnote2 Let us also recall that the forces and relations of production, although distinct realities, never exist separately, but always combined in a specific manner. ‘Modes of production’ or ‘social forms of production’ are the various specific forms of such combination. Let us look more closely at these definitions, which call for some comment.

First of all, the productive forces include what I have called the ‘intellectual’ means of transforming nature. By this I mean all the ‘knowledge’ of nature which a society can have, together with the totality of technical procedures, rules for manufacture of tools, rules for use of the body in work, etc. We see that at the heart of the most material relation between man and the material reality which surrounds him, there is to be found a complex ensemble of representations, ideas, schemas, etc., which I shall term ‘ideal’ realities, and whose presence and intervention is needed for material activity to take place. Nowadays, anthropology has undertaken an inventory of these ideal realities contained in the various material processes of the societies it analyses. This is the vast domain of ethnoscience, which studies native taxonomies of plants, animals, the climate, the soil, rules for the production of tools, etc. This is also the object of the studies of science and technology undertaken by such historians as Joseph Needham—with respect to China—or André Haudricourt.footnote3